🌍 Hi all, I'm Luke Stanley. ⌨️ I started coding when I was 12 on a BBC Micro. I wanted to play games, but the disk drive didn't work, so I learned BASIC. 💻When I got an IBM PC, I was fascinated by Visual Basic's RAD tools. I explored the Win32 API, ActiveX, OLE (interesting component systems), multimedia, and networking—teaching myself as much as I could. 💿 There's a big gap between the rich multimedia learning tools I explored as a child—like Encarta, Ancient Lands, and The Way Things Work CD-ROMs—and how we can explore our personal, internal worlds with computers. 🎨 It inspires me to explore visual language, hypertext, information retrieval, generative technology, semantic desktops, and self-assembling knowledge graphs to bridge that gap—something I'm still working on. 🤖 At 16, I got into what we'd now call machine learning. Inspired by Paul Graham's "A Plan For Spam" and my curiosity about Myers-Briggs categories, I created a dataset to predict introversion vs. extroversion from text samples using POPFile (a Bayesian spam filter) repurposed for a personality classification task called "Mindprise." Around this time, I first interacted with futurists and AI safety figures like Eliezer Yudkowsky via IRC chat. I became an early supporter of the Singularity Institute (now MIRI) and learned about them through a memetics group on the same IRC server. 🦁🤝 I met Lion Kimbro through an online BBC Magazine article, and we became friends. Lion got me into Carl Sagan, The Great Story, and making visual language lifemaps. Around 2008, after Lion showed me his lifemap, I started the

ThoughtTrail

project, a metadata knowledge graph plugin system powering a contextual search tool that identified conversation topics in MSN Messenger and Skype chats. It displayed Google Images and YouTube videos in a sidebar, with Wikipedia information accessible via a Python plugin. Worried about privacy issues, I pivoted to exploring a Visual Language Wiki (Cavepaint). I’m still interested in semi-automatic timelines and knowledge outline generation. 👣 I'm really intrigued by @Maggie Appleton's "*Barefoot Developers*" concept, I want to see spreadsheets, databases, and apps accessible and interoperable for non-developers. I want my 70-year-old mum, who isn't technical, to create tools for the cat charity she supports. Helping people move beyond basic tables to databases and apps they can share, understand, and control is important and will unlock a lot. I agree with Alan Kay: the information revolution has barely started. 🔎 I built a search tool for my ChatGPT conversation dump containing over 4000 conversations. 🗺️ @Amelia Wattenberger's visual representations showing the same data as lists, tables, grids, maps, I think with D3 and embeddings. 🤔 I'm curious about personal AI assistants for meaning prediction and emotional insight—what I think is important and why. 👨‍💻 Daniel Harris introduced me to the Future of Coding group, who I think was introduced by Steve Ruiz of Tldraw. More recently, I connected with @Kartik Agaram via Lion’s Internet Office on Discord. 🥺 We should map out probable suffering and its remediation, improve our collective intelligence, and be mindful of the Fermi Paradox—treading carefully towards Pandora's box. Some interests include: 📘 Restricted vocabulary explanations like Thing Explainer by XKCD 🔗 Improving representation and connectivity in embedding spaces 🤖 Fine-tuning small models for explanations, and on my ChatGPT dump to predict what I might say! 🌀 Exploring deep state-space models, training continuation, small models, and constraint solvers 📥 Addressing information overload with unified systems 🗺️ Automatically creating visual maps, emoji topic outlines, ontologies, and hierarchical knowledge graphs. I made a Paul Graham essay map using embeddings 🧡 We Feel Fine by Jonathan Harris 🌐 Map-making to connect personal understanding with broader knowledge, and developing semi-automatic timelines 💥 I'm curious what high-impact solutions there are for big problems, and just having fun playing and exploring.